Research map / Detection and segmentation / Open-set and few-shot detection
BusterNet: Detecting Copy-Move Image Forgery with Source/Target Localization
Yue Wu, Wael Abd-Almageed, Prem Natarajan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the manipulation branch (two-branch design, direct training) (ablation) | BusterNet · Synthetic 10K test set | recall target: 38.97% -> 47.37% | Improved |
| add the manipulation branch (two-branch design, direct training) (ablation) | BusterNet · Synthetic 10K test set | recall source: 32.28% -> 34.12% | Improved |
| add stage-wise training with auxiliary tasks (ablation) | BusterNet · Synthetic 10K test set | recall source: 34.12% -> 41.64% | Improved |
| add stage-wise training with auxiliary tasks (ablation) | BusterNet · Synthetic 10K test set | recall target: 47.37% -> 53.61% | Improved |
| add stage-wise training with auxiliary tasks (ablation) | BusterNet · Synthetic 10K test set | accuracy: 92.74% -> 93.02% | Improved |
| add fusion and end-to-end fine-tuning to the Simi-Det branch (ablation) | BusterNet · CASIA CMFD | pixel/image AUC: AUC improves by 3-4% | Improved |
| add fusion and end-to-end fine-tuning to the Simi-Det branch (ablation) | BusterNet · CASIA CMFD | image-level F-score: 75.85% -> 75.98% | Improved |
| add fusion and end-to-end fine-tuning to the Simi-Det branch (ablation) | BusterNet · CASIA CMFD | pixel-level F-score (protocol A): 59.16% -> 67.05% | Improved |
| use BusterNet instead of dense-field method [9] | copy-move detector · CoMoFoD no attack | pixel-level F1 (protocol B, overall): 41.83% -> 49.26% | Improved |
| use BusterNet instead of dense-field method [9] | copy-move detector · CoMoFoD no attack | F1 on correctly detected subset: 87.82% -> 80.09% | Hurt |
| apply severe JPEG compression (JC1, quality 10) attack (ablation) | BusterNet · CoMoFoD | correctly detected images: 117 -> 60 | Hurt |
| restrict source/target scoring to opt-in samples | BusterNet · source/target discernibility | accuracy: 11.83% -> 77.49% | Improved |
Later papers that compared with this method
1 later accepted paper reports BusterNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- DOA-GAN: Dual-Order Attentive Generative Adversarial Network for Image Copy-Move Forgery Detection and Localization (CVPR 2020 · better on 3 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FakeShield: Explainable Image Forgery Detection and Localization via Multi-modal Large Language Models (ICLR 2025 · similarity 0.87)
- ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and Localization (NeurIPS 2025 · similarity 0.86)
- Reinforced Multi-teacher Knowledge Distillation for Efficient General Image Forgery Detection and Localization (AAAI 2025 · similarity 0.86)
- Learning Discriminative Noise Guidance for Image Forgery Detection and Localization (AAAI 2024 · similarity 0.85)
- BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation (NeurIPS 2026 · similarity 0.85)
- Exposing the Deception: Uncovering More Forgery Clues for Deepfake Detection (AAAI 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
